The purposes of this study were first to develop a system which statistically tests results of radiotherapy and which automatically acquires an optimal discrete-valued attribute by dividing and grouping continuous-valued attributes, and second to find the optimal range of values such as tumor dose by taking account of the conditions and statistics in ROGAD (Radiation Oncology Greater Area Database), a multi-institutional database in Japan. Our ultimate goal is to assist clinical decision making for every patient. In this research, two algorithms for acquiring a boundary value were developed without detecting false boundaries or accidental errors of acquired boundaries. The resolution of detected discrete-valued attributes and speed of convergence were confirmed to be practical. The optiin' al range of given tumor dose with the best reaction and with the fewest complications is expected to be clarified.

View full abstract

A radiation accident occurred at a medical linear accelerator facility under construction in Japan. The radiation source was a 3-and 6-MV potential drop accelerator designed to produce X-rays for radiation therapy. This accelerator was also capable of producing a 5tol4-MV swept electron beam. During setting up, an operator turned on the accelerator to test the beam not knowing that a man was working on the ceiling above the accelerator. Thus, an X-ray beam was emitted against the ceiling and the man was exposed to 10-MV of X-ray irradiation. However, no obvious physical symptoms were noted. Dose estimation was made from reconstruction of the accident and clinical examinations including chromosome analysis. Mean dose of the whole body ranged from 70 to 180 mSv. Estimated dose from his right foot to hand was between180 to 900 mSv.

View full abstract

We previously reported a method of pixel value averaging for microcalcification specks on digitized mammographic phantom images. These average values are considered to represent the pixel value of a speck for quality control in comparison to its maximal peak value. In this report, we described an automatic processing procedure using a spreadsheet to derive the average and peak values of the pixel value of a speck. A hundred and twenty specks in 10 phantoms were processed using the spreadsheet, and average and peak values were compared. In the examination of the digitized mammography images of RMI 156 phantoms for two calcification speck groups, the average and peak values showed good correlation. Therefore, we concluded the peak values could be used as substitutes for the average values for a quality control purpose. However, as the regression coefficients of the two speck groups were significantly different, the relationship between the average and peak value seemed to depend on speck sizes.

View full abstract